Equation of state, phase stability, and amorphization of
SnI[superscript 4] at high pressure and temperature
The MIT Faculty has made this article openly available. Please share
how this access benefits you. Your story matters.
Citation
Grocholski, Brent, Sergio Speziale, and Raymond Jeanloz.
“Equation of state, phase stability, and amorphization of SnI_{4}
at high pressure and temperature.” Physical Review B 81.9
(2010): 094101. © 2010 The American Physical Society
As Published
http://dx.doi.org/10.1103/PhysRevB.81.094101
Publisher
American Physical Society
Version
Final published version
Accessed
Wed May 25 21:38:54 EDT 2016
Citable Link
http://hdl.handle.net/1721.1/58624
Terms of Use
Article is made available in accordance with the publisher's policy
and may be subject to US copyright law. Please refer to the
publisher's site for terms of use.
Detailed Terms
PHYSICAL REVIEW B 81, 094101 共2010兲
Equation of state, phase stability, and amorphization of SnI4 at high pressure and temperature
Brent Grocholski*
Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge,
Massachusetts 02130, USA
Sergio Speziale
Helmholtz Centre Potsdam, GFZ German Research Centre for Geosciences, Section 3.3, Telegrafenberg, D 228,
D-14473 Potsdam, Germany
Raymond Jeanloz
Department of Earth and Planetary Science, University of California–Berkeley, 307 McCone Hall, Berkeley, California 94720-4767, USA
共Received 3 November 2009; revised manuscript received 20 January 2010; published 2 March 2010兲
We have measured the pressure-induced amorphization of tin iodide 共SnI4兲 as a function of temperature and
found a small 共⬍1 GPa兲 increase in the initiation pressure of the transition from 293 to 523 K. We have also
determined the P-V equation of state while the material undergoes the amorphization, including multiple
compressions of a single sample, using a diamond-anvil cell and x-ray diffraction. Apparent stiffening upon
multiple compression of the sample can be explained by either an increase in nonhydrostaticity during loading
or, possibly, changes in the initial crystal structure upon decompression recrystallization. Contrary to previous
interpretations of the amorphization being a two-stage process, with SnI4 converting to a second 共unidentified兲
crystalline phase before the loss of Bragg intensities, the present experiments show that it is also possible to
amorphize SnI4 without the appearance of the second crystalline phase. While the new diffraction peaks appear
upon the first compression of a sample, subsequent compressions of the same sample do not show these new
peaks; instead, long-range ordering of the original cubic crystal structure is lost.
DOI: 10.1103/PhysRevB.81.094101
PACS number共s兲: 62.50.⫺p, 81.40.Vw, 61.05.cp, 64.70.P⫺
I. INTRODUCTION
Tin iodide 共SnI4兲 is a van der Waals-bonded molecular
solid with a cubic structure at atmospheric pressure. The
crystal structure at room pressure and temperature, determined by Dickenson1 and further refined by Meller and
Fankuchen,2 can be described as containing SnI4 tetrahedra
arranged in a face-centered-cubic lattice.3 Under compression at room temperature at around 15 GPa, SnI4 undergoes
reversible solid-state amorphization that was first recognized
using x-ray diffraction4 and Raman scattering.5 This transition was initially thought to involve only the original cubic
crystal phase 共CPI兲 but this interpretation has been challenged by the observation of a second crystal phase 共CPII兲 at
about 7.5 GPa.6 While several theories have been proposed
to account for the loss of Bragg intensity with pressure,7,8 the
idea of pseudomelting7,9 has been invoked specifically to account for the amorphization of tin iodide with pressure.
Amorphization of tin iodide is accompanied by a large
increase in electrical conductivity, first noticed by Riggleman
and Drickamer.10 Metallization, which is achieved by 15
GPa, requires this weakly bonded material to generate
enough orbital overlap for electron delocalization in order to
account for the electrical conductivity increase with pressure.
Mössbauer spectroscopy using 119Sn and 129I suggests that
the material may be polymerizing, forming -共SnI4兲x- chains
of random length that account for both the amorphous character of the sample and the increase in electrical
conductivity.3 This interpretation is supported by additional
Mössbauer and conductivity measurements in the diamond
cell.11 However, the appearance of CPII happens to occur at
the same pressure as the increase in conductivity and this
1098-0121/2010/81共9兲/094101共7兲
correspondence suggests that metallization may be due to
CPII 共Ref. 6兲 rather than amorphization of the sample.
We have conducted a set of experiments to determine the
change in amorphization pressure with temperature and measured the equation of state at room temperature for pristine
and multiply recompressed samples. We have found that, in
certain cases, the appearance of CPII is not necessary in
order to amorphize the sample. In addition, we have crystallized a new phase of tin iodide at low pressure 共2.7 GPa兲
through sudden decompression of the sample at temperatures
above 573 K.
II. EXPERIMENTAL METHOD
Nine samples of SnI4 共Alfa Aesar, 99.998% purity兲 were
loaded in rhenium gaskets and compressed in a diamondanvil cell to roughly 20 GPa in order to fully amorphize each
sample. We used a membrane cell and diamonds with
350 ␮m culets. The sample chamber was roughly 100 ␮m
in diameter and 15– 25 ␮m thick. Pressure was measured
using the equation of state of gold12 with either a gold flake
共samples 1–5兲 or a ⬃60: 40 mixture of SnI4 to gold powder
共samples 6–9兲. The gold was used as an internal pressure and
intensity standard. Powder x-ray diffraction patterns were
taken at approximately 1 GPa pressure intervals, using a 25
keV x-ray beam of ⬃20 ␮m diameter at the Advanced Light
Source Beamline 12.2.2. Data were collected for
⬃100– 200 s with a MAR-345 image plate 共Fig. 1兲.
Heating was performed using a ring heater positioned inside the cell body. The diamond-anvil cell was placed in a
container flushed with inert gas 共argon with 5% hydrogen兲.
094101-1
©2010 The American Physical Society
PHYSICAL REVIEW B 81, 094101 共2010兲
GROCHOLSKI, SPEZIALE, AND JEANLOZ
7
8
9
1 0
1 1
2 q
1 2
1 3
1 4
1 5
1 6
1 7
1 8
(B )
1 9
5
6
7
8
9
1 0
1 1
2 q
1 2
1 3
1 4
1 5
1 6
1 7
1 9
1 0 0
1 0 0
3 0 0
A z im u th
A z im u th
2 0 0
3 0 0
6 .2 G P a
8
9
1 0
1 1
2 q
1 2
2 q
1 3
1 4
1 5
1 6
1 7
1 8
1 9
5
6
7
8
9
1 0
1 1
1 2
1 3
1 4
1 5
1 6
1 7
1 8
1 9
1 0 0
A z im u th
3 0 0
3 0 0
A z im u th
2 0 0
1 0 0
0
7
2 0 0
6
5 .9 G P a
0
5
1 8
0
6
0
5
2 0 0
(A )
7 .9 G P a
8 .3 G P a
FIG. 1. X-ray diffraction intensity as a function of scattering angle 2␪ 共in degrees兲 and azimuth for SnI4 at comparable pressures for two
different samples: first compression of a sample at 293 K 共sample 6兲 on the left and fourth compression at 373 K 共sample 5.4兲 on the right.
Three new diffraction lines appear as spots on the left 共arrows兲. Spottiness is absent on the right and the diffraction lines are smooth due to
a broad distribution of crystal grain orientations relative to the x-ray beam. No new diffraction features appear in this sample above 7 GPa.
In addition, the 311 reflection to the left of the highest intensity 222 line 共2␪ ⬃ 9兲 is also absent.
The container design allowed the x-ray beam to reach the
sample, but made optical access impossible, preventing the
use of ruby fluorescence for pressure calibration. The temperature was measured using a thermocouple that was placed
within 1 mm of the diamond tips. The temperature inside the
ring heater had been previously calibrated but no further attempts were made to measure the temperature in situ. While
the temperature measurement has 1 ° – 2° precision, the absolute accuracy of the temperature measured in experiments
is only ⬃25°. This is in part due to insufficient and irregular
thermal insulation of the ring heater from the diamond backing plates.
Sample 1 was cryogenically loaded with an argon pressure medium but this was found to be unnecessary because
tin iodide is a soft material 共K0 ⬃ 11 GPa兲 and does not appear to benefit from the presence of a quasihydrostatic pressure transmitting medium. Therefore, subsequent samples
共2–9兲 were loaded without argon. A separate sample 共sample
10兲 was loaded at room temperature and compressed a total
of two times to assess the nonhydrostatic pressure gradient
using the ruby fluorescence technique.13 The pressure variation was found to be at most 0.5 GPa at 20 GPa, for both the
first and second compressions of the sample. This is consistent with previously reported measurements of pressure
variations across SnI4 samples loaded inside gasketed diamond cells.4,14 Most of the samples in this experiment were
taken to high temperature, which also reduces pressure gradients due to nonhydrostaticity 共Table I兲.
III. RESULTS
A. Equation of state
We compressed nine different samples of tin iodide a total
of 13 times. Figures 1共A兲 and 2共A兲 show an example, with
the patterns illustrating the appearance of the new phase CPII
at roughly 7.5 GPa as noted by Hamaya et al.6 Most of the
intensity of the new peaks comes from individual spots that
are easiest to see in Fig. 1共A兲 in the pattern collected at 7.9
GPa. Most of the samples that we compressed showed the
same behavior between 6 and 8 GPa. Figure 1共B兲 shows
diffraction patterns from a sample that had been pressure
amorphized and recrystallized by decompression three times
before recompression at 373 K 共sample 5.4兲. The patterns are
094101-2
PHYSICAL REVIEW B 81, 094101 共2010兲
EQUATION OF STATE, PHASE STABILITY, AND…
TABLE I. Summary of all experimental data runs. The second column gives the range of pressure over which the Birch-Murganhan
equation of state was fit. Parameters 共K0T , K0T
⬘ 兲 are given for both third and second-order fits for room-temperature samples. Hightemperature samples were not fit due to poor constraint of V0. Amorphous transition pressures in general are estimates based on the intensity
decrease in the 440 reflection with pressure and the general resolvability of the diffraction pattern as a whole. The transition pressure for
sample 5.2–5.5 were determined by fitting the decrease in intensity of the 440 reflection and taking the 50% value using Eq. 共1兲 共Fig. 4兲.
Sample No.
P
T
共K兲
1
2
3
4
5.1
2.4–15.0
4.0–14.7
2.4–19.5
7.7–21.7
0.2–16.9
293
348
398
448
523
5.2
5.3
5.4
5.5
6
7
2.1–17.6
0.1–17.0
5.6–17.3
0–17.8
0.6–16.7
3.2–13.7
488
448
373
293
293
573
8
3.2–16.6
⬎573
9
1+6
0–3.0
0–16.7
598
293–573
K0T
K0T
⬘
ⴱ
K0T
12.2共5兲
3.5共2兲
10.6共1兲
13.5共4兲
8.9共4兲
4.2共3兲
5.4共2兲
Pt
Comments
⬃12– 15
⬃13
⬃15– 19
⬃17– 19
⬃17– 19
Argon loaded
CPII absent
Much higher Pt than other samples
Multiple compressions, last 4 not showing any evidence of
CPII;
⬃14.2
⬃15.4
⬃14.5
⬃14– 15
⬎13.7
14.2共1兲
10.6共1兲
V0 = 12.12共2兲
Gasket failure CPII appears ⬃5.3 GPa
Gasket failure, new phase appears upon decompression to
2.3 GPa. No CPII with second compression
Loses 311 and 331 diffraction lines upon decompression
from 3 GPa
V0 = 12.261共1兲
⬎14– 16
10.4共3兲
4.1共1兲
10.6共1兲
S a m p le 5 , 1 s t c o m p r e s s io n
S n I4
S a m p le 5 , 5 th c o m p r e s s io n
S n I4
1 9 .5 G P a
1 6 .9 G P a
1 6 .9 G P a
1 5 .8 G P a
1 5 .9 G P a
In te n s ity
In te n s ity
2 0 .5 G P a
1 4 .0 G P a
1 0 .6 G P a
1 0 .7 G P a
6 .4 G P a
6 .7 G P a 6 2 2
A u
2 0 0
4 4 4
3 3 2
4 4 0
0 .3 G P a
2 .0
A u
1 1 1
3 2 1
3 3 1
3 0 2
4 4 4
3 1 1
2 2 2
4 2 1
2 .5
3 .0
4 0 0
A u
2 0 0
4 0 0
6 2 2
1 .5
1 3 .8 G P a
3 .5
4 4 0
0 G P a
4 .0
1 .5
d ( )
2 .0
3 3 2
A u
1 1 1
3 2 1
4 2 0
2 2 2
4 2 1
2 .5
3 0 2
3 .0
3 .5
4 .0
d ( )
FIG. 2. Integrated x-ray diffraction for sample 5 at various pressures. The patterns on the left are for the first compression 共at 523 K兲
while the patterns on the right are from the fifth compression of the same sample after four decompression crystallizations from the
amorphous state. Arrows indicate the appearance of CPII in the diffraction pattern. The lack of CPII in the right pattern makes it easier to
extract the positions of diffraction lines: especially the 622 and 444 reflections. The sample amorphizes in both cases at roughly the same
pressure.
094101-3
PHYSICAL REVIEW B 81, 094101 共2010兲
GROCHOLSKI, SPEZIALE, AND JEANLOZ
S a m
S a m
S a m
S a m
S a m
S a m
S a m
S a m
S a m
fit to
1 8 0 0
1
2
3
4
5 .1
5 .2
5 .3
5 .4
5 .5
3 K d a ta
S a m p le 6
S a m p le 7
S a m p le 8
S a m p le 9
L y n c h a n d
D r ic k a m e r ( 1 9 6 6 )
c o m p u ta tio n
e x p e r im e n t
F u c h iz a k i e t a l. ( 2 0 0 0 )
V o lu m e ( 
3
)
1 6 0 0
p le
p le
p le
p le
p le
p le
p le
p le
p le
2 9
1 4 0 0
1 2 0 0
1 0 0 0
0
5
1 0
P re s s u re (G P a )
1 5
2 0
FIG. 3. Pressure-volume plot for all data. The data in gray are
from previous studies. The volumes were calculated using 2 to 12
reflections of the cubic crystal structure. The large deviation in
Lynch and Drickamer 共Ref. 16兲 at high pressures may be due to a
misidentification of the amorphous humps as the 222 and 622 reflections. All our pressures were determined using gold as an internal standard.
much smoother than those of Fig. 1共A兲 due to smaller and
more randomly oriented grains that form during recrystallization, along with no indication of CPII formation. All diffraction patterns collected from sample 5 共runs 5.2–5.5兲 have
a similar smoothness and absence of the CPII diffraction
lines.
The integrated x-ray diffraction patterns are shown in Fig.
2 from a sample collected at 523 K 共sample 5.1兲 upon its
initial compression and the same sample at 293 K 共sample
5.5兲 during its fifth compression. The 0 GPa diffraction patterns from the recrystallized samples 共5.2–5.5兲 are similar,
but not identical, to the pattern of the initial sample: the 311
and 331 diffraction lines are missing and the 420 reflection
appears in these recrystallized samples. The lattice parameter
of the unit cell also appears to be ⬃1.1% smaller 关a0
= 12.12共2兲 Å兴 than our room-pressure value of 12.261共1兲 Å,
after compressing the sample four times. The change in intensity of lattice reflections is perplexing, as 420 is not a
forbidden reflection but has zero intensity as long as the SnI4
tetrahedra are not distorted. Tetrahedral distortion can decrease the 311, 331 intensities to zero, but the 420 reflection
should be accompanied by the 411 reflection—which is not
observed in this sample. This behavior is not limited to
sample 5, as the same reflections are lost in sample 9 upon
catastrophic decompression from 3 GPa to room pressure at
598 K without any amorphization. Sample 8 has the 420
reflection but also contains the 311, 331, and 411—indicating
that tetrahedral distortion is possible.
Our measurements of unit-cell volumes are consistent
with values reported by Fuchizaki et al.15 and up to 15 GPa
with the data of Lynch and Drickamer16 共Fig. 3兲. Above 15
GPa, the low volumes from Lynch and Drickamer are probably due to misidentification of the amorphous humps as the
peak positions of the 222 and 622 reflections. Background
increases due to pressurization and amorphization combined
with apparent peak broadening from the appearance of CPII
make different hkl lattice reflections become unresolved at
different pressures. Volumes are calculated from 8 to 12 reflections at low pressure and 2 to 4 reflections at high pressure. Most of the volumes in Fig. 3 fall within a few percent
of the average volume for any given pressure. The scatter is
due in part to the high compressibility of the material—
errors in pressures can lead to large relative errors in volume.
Many of the deviations appear to be systematic and potentially dependent on the initial conditions of the sample or the
specific pressure distribution in the sample chamber.
A third-order isothermal Eulerian finite-strain 共BirchMurnaghan兲 equation of state12 was used to fit the volumes
collected from samples 1 and 6 at room pressure, with a V0
of 1843.2 Å3 and the recompressed sample 5.5 was fit with
a resulting V0 of 1780.4 Å3. The values for the bulk modulus
and pressure derivative are K0T = 10.6共1兲 GPa and K0T
⬘
= 4.1共1兲 共subscripts zero and T indicate zero pressure and
isothermal conditions, respectively兲. Fitting with a fixed
K0T
⬘ = 4 共corresponding to a second-order Birch equation of
state兲 gives a bulk modulus of 10.6共1兲 GPa. The recompressed sample volumes increasingly deviate from the measured first-compression volumes with each compression. The
divergence occurs within the range at which the CPII lines
appear in the reference data and may reflect lack of stress
共pressure兲 relaxation across our sample and between the various phases present 共including the pressure calibrant兲. The
result after four compressions is an increase in bulk modulus
at room temperature to K0T = 13.6共1兲 GPa with K0T
⬘ = 4.2共3兲
or K0T = 14.2共1兲 GPa with K0T
⬘ = 4. At pressures above 15
GPa all of the isothermal compression curves collapse
back on each other when all of the samples are mostly amorphous.
B. Pressure-induced amorphization
Figure 2 shows a comparison between amorphization with
and without the appearance of CPII in the integrated diffraction pattern. The diffraction patterns in Fig. 2共A兲 are similar
to those previously published,4,6 with the location of the CPII
diffraction lines shown as arrows on the 10.6 GPa trace. The
sample eventually amorphizes at 13–15 GPa, although determining the point at which this happens is difficult due to the
appearance of the CPII diffraction lines. The traces shown in
Fig. 2共B兲 demonstrate the sample can become amorphous
without the appearance of a second 共CPII兲 crystal phase. In
this case, it becomes much easier to quantify the amorphization pressure. We quantify how the intensity of the 440 diffraction line of SnI4 changes with pressure, relative to the
111, 220, or 311 gold lines, because its intensity correlates
well with the overall change in diffraction pattern 共see Fig.
2兲 and it is the last peak to lose intensity upon amorphization.
The results are fitted to Eq. 共1兲,
再
冎
P − Pt
Isample
= 0.5 ⫻ 1 − tanh
.
Icalibrant
wt
共1兲
The amorphous transition pressure 共Pt兲 is the point at which
the normalized relative intensity is at 50% and wt is a mea-
094101-4
PHYSICAL REVIEW B 81, 094101 共2010兲
EQUATION OF STATE, PHASE STABILITY, AND…
1 .0
R e la tiv e In te n s ity
0 .8
le 5
le 5
le 5
le 5
le 6
le 7
s a m
s a m
s a m
s a m
s a m
.2 (2 1 5
.3 (1 7 5
.4 (1 0 0
.5 (2 0 C
(2 0 C )
(3 0 0 C )
p le 5 .2
p le 5 .3
p le 5 .4
p le 5 .5
p le 6
1 6 .0
C )
C )
C )
)
1 5 .5
T r a n s itio n P r e s s u r e ( G P a )
S a m p
S a m p
S a m p
S a m p
S a m p
s a m p
F it to
F it to
F it to
F it to
F it to
0 .6
0 .4
0 .2
t o s a m p le 5
S a m p le 5
S a m p le 6
L in e a r F it
1 5 .0
1 4 .5
1 4 .0
1 3 .5
0 .0
1 3 .0
5
1 0
1 5
P re s s u re (G P a )
2 0
3 0 0
2 5
FIG. 4. Normalized intensity as a function of pressure for the
共440兲 reflection. The data have been fit using Eq. 共1兲 and the parameters for the fits are in Table II. The normalized intensities from
samples 6 and 7 are plotted for comparison—there appears to be an
initial decrease in the relative intensity due to the appearance of
new diffraction lines in the sample. This makes it difficult to define
the initial intensity and therefore quantify the amorphous transition
pressure. We fit the high-pressure 共⬎10 GPa兲 data for sample 6 by
rescaling the zero-pressure relative intensity. Sample 7 does not
appear to have undergone an amorphization and upon sudden decompression from ⬃13 to ⬃4 GPa 共arrow兲, recovers a reasonable
relative intensity of ⬃1.0.
sure of the width of the amorphization 共Fig. 4兲. In order to
minimize the effect of pressure-induced amplitude reduction,
the 100% value is taken as the average of the sample to
calibrant ratios of data points ranging from 4 to 6 GPa.
Samples 6 and 7 are plotted to show the difference between
how the intensity of the 440 line falls off as a function
of pressure when the new diffraction lines 共CPII兲 appear
in the sample. Sample 6 was fit using a renormalized
form of Eq. 共1兲, as well, although due to the appearance
of 共CPII兲 we need to take the 100% intensity value at
11 GPa instead of lower pressure. The transition pressure
as a function of temperature 共or compression兲 is given in
Table II. The error on each value of Pt was determined by
differencing the pressure at which the intensity was 40%
and 60% of the original intensity. These data are plotted
in Fig. 5 and a least-squares fit to the transition pressures gives a slope of 3.3共⫾4.1兲 MPa/ K and an interTABLE II. Parameters from fitting sample 5 in Fig. 4 with Eq.
共1兲.
Compression
Pt
共GPa兲
wt
共GPa兲
2
3
4
5
15.5共0.4兲
14.2共0.8兲
15.4共0.6兲
14.5共0.8兲
2.2共0.4兲
3.9共0.3兲
3.9共0.3兲
4.2共0.2兲
3 5 0
4 0 0
T e m p e ra tu re (K )
4 5 0
5 0 0
FIG. 5. Transition pressure as a function of temperature as determined by fits to the data in Fig. 4. Sample 6 was collected at 293
K but has been shifted to 296 K to make it easier to see the error on
the measurement. The data from sample 5 were fit using linear
regression. Error bars were determined by taking the difference between the transition pressure and the pressure at which the normalized intensity was 40% or 60%.
cept of 13.7共⫾1.8兲 GPa. In all cases, these are amorphization pressures obtained by increasing the pressure of the
sample.
Amorphization pressures were estimated qualitatively for
the different sample loadings that have the CPII phase and
are listed as a range in Table I. In general, Pt for different
samples appears to be load dependent, ranging from a low of
⬃13 GPa for sample 2 to a high of 17–19 GPa for sample 4.
Our results are consistent with Hamaya et al.’s6 findings that
document a large hysteresis in the amorphization transition,
with the reverse transformation 共amorphous→ crystalline兲
occurring below 1.8 GPa.
IV. DISCUSSION
Over the course of 13 sample compressions and roughly
300 diffraction patterns, the most consistent observation was
the loss of diffraction intensity that documented the appearance of an amorphous material. These high-pressure amorphous solids recrystallized upon pressure release, generating
crystals with smaller grain size than the initial sample. The
amorphization takes place regardless of temperature or the
appearance of CPII. The repeatable amorphization combined
with the tendency to measure uniform volumes when the
sample is mainly amorphous indicates that much of the scatter in the volume data and irreproducibility in the number of
crystal phases in the sample is dependent on the specifics of
the sample stress environment 共and perhaps history兲. The
weak intermolecular forces in tin iodide may make it sensitive to even small deviatoric stresses, which can change the
amorphization pressure of crystals.17 At pressures above 15
GPa, all samples appear to behave about the same, apparently due to a more uniform stress environment from sample
to sample.
094101-5
PHYSICAL REVIEW B 81, 094101 共2010兲
GROCHOLSKI, SPEZIALE, AND JEANLOZ
There are several possible reasons for the differences in
the amorphization pressure between different samples and
sample loadings. If we treat the temperature dependence of
the transition pressure as that of a first-order transition we
can define a Clayperon slope as Eq. 共2兲,
6 2 2
3 0 2 4 0 0
共2兲
4 4 4
3 2 1
2 1 0
2 1 1
3 1 1
where the change in volume should be negative. The data
collected for sample 5 and shown in Fig. 5 has a slightly
positive slope of 3.3共⫾4.1兲 MPa/ K that implies little or no
change in entropy between CPI and the amorphous state.
This is consistent with the interpretation of the pressureinduced amorphization of calcium hydroxide.18 The variation
in amorphization pressures among our samples may also be
due to subtle differences in the deviatoric stresses between
samples. Catalli et al.17 have found the pressure-induced
amorphization of Ca共OH兲2 is likely facilitated by nonhydrostatic 共deviatoric兲 stress and they do not observe this transition in a quasihydrostatic pressure medium 共argon兲. Although
the nonhydrostatic stresses in our system appear to be relatively low 共⬍0.5 GPa at 20 GPa兲, tin iodide may be extremely sensitive to even small deviatoric stresses that
change the amorphization pressure by up to several
gigapascals.19
The pressure-volume curve in Fig. 3 shows that, although
we have a fairly consistent general trend, significant deviations appear that are not easy to explain without considering
nonhydrostatic effects. In the case of sample 5, it appears
that the recrystallization does not produce quite the same
material as we started with. After each decompression cycle
the sample appears to have smaller unit-cell volumes and,
possibly, slightly distorted tetrahedra. This results in less
compressible crystals 共i.e., an increase in the bulk modulus兲.
These types of deviations are likely due to the sample
environment 共i.e., deviatoric stress, grain size, and internal
strains兲 as the uniformity of the measured volumes above 15
GPa is due to all samples being primarily amorphous and
presumably reflect a similar sample environment between
different loadings. Figure 6 shows an example where shear
共via gasket blowout兲 generates a new 共presumably metastable兲 phase of tin iodide. This brings up the possibility that
CPII is a shear-induced phase as well that can coexist with
CPI until the sample becomes disordered and it converts to
an amorphous solid, either directly or via back-conversion to
CPI as is seen in Fig. 6.
Regardless of the nature of the CPI-CPII phase boundary,
there is no need for the sample to go through the CPII
phase for amorphization. The production of the amorphous
phase from two different crystal structures seems to undermine the pseudomelting hypothesis as suggested by
Fuchizaki et al.20 The two materials that have been cited as
examples of pseudomelting are water-ice9 and quartz,21
and both are thought to have the negative dTm / dP melting
slopes necessary for the melt curve to return to low
temperature: a requirement for explaining room-temperature
amorphization according to this model. That the melting
slope of SnI4 is positive, according to molecular-dynamics
simulations15,20,22,23 and direct measurement,23 seems to
4 1 1
s a m p le 8
0 .9 G P a
4 2 1
4 2 0
3 3 1
In te n s ity
dPt ⌬S
=
,
dT ⌬V
4 4 0
2 2 2
s a m p le 8
2 .7 G P a
A u 2 0 0
A u 1 1 1
4 4 0
2 2 2
s a m p le 5 .5
2 .4 G P a
3 2 1
2 1 0 2 1 1
6
6 2 2
4 0 0
3 0 2
8
4 2 0
1 0
4 2 1
4 4 4
1 2
2 q (d e g re e s )
1 4
1 6
1 8
FIG. 6. Comparison of data collected at similar pressures
共⬃2.5 GPa兲 and at 0.9 GPa 共sample 8兲 as an example of a mixedphase diffraction pattern. Sample 8 at 2.7 GPa is missing all of the
major diffraction lines. The major lines are all present in the mixedphase pattern along with a number of reflections 共i.e., 411 and 420兲
that have zero intensity for the SnI4 structure provided the tetrahedra are regular.
eliminate this hypothesis. Moreover, pseudomelting may
not explain high-pressure amorphization, in general, as
many systems produce an amorphous state appearing like
a highly disordered crystal24 or exhibiting a memory of
the crystal phase from which it came.25 Assuming that both
paths to amorphization 共CPI→ CPII→ amorphous or CPI
→ amorphous兲 are possible, this would support the idea that
mechanical destabilization due to polyhedral disordering
causes loss of long-range order, such as proposed by Tse and
Klug26 for SiO2. Such disordering may well be consistent
with the idea of SnI4 chains being formed, as well, resulting
in the observed increase in electrical conductivity.3
V. CONCLUSION
We have conducted a set of experiments to determine the
change in amorphization pressure with temperature and have
measured the equation of state of SnI4 at room temperature.
We have found that the appearance of a second crystal phase
is not necessary in order to amorphize the sample. More
generally, our results seem inconsistent with the pseudomelting hypothesis, instead favoring other mechanisms of amorphization. Since both CPI and CPII amorphize eventually, it
seems likely that there is at least one additional thermodynamically stable 共crystalline兲 phase at high pressures. The
possibility also remains that the glass formed is the stable
phase, especially in light of a zero or positive Clayperon
094101-6
PHYSICAL REVIEW B 81, 094101 共2010兲
EQUATION OF STATE, PHASE STABILITY, AND…
slope. Finally, we have also observed a new, shear-induced
metastable phase of tin iodide, obtained by sudden partial
decompression of the sample from high pressure.
ACKNOWLEDGMENTS
The authors would like to thank Simon Clark, Martin
Kunz, Sander Caldwell, and A. E. Gleason for experimental
assistance. We would like to thank Dave Walker for the con-
*b.grocholski@gmail.com
Dickenson, J. Am. Chem. Soc. 45, 958 共1923兲.
2 F. Meller and I. Fankuchen, Acta Crystallogr. 8, 343 共1955兲.
3
M. Pasternak and R. D. Taylor, Phys. Rev. B 37, 8130 共1988兲.
4
Y. Fujii, M. Kowaka, and A. Onodera, J. Phys. C 18, 789
共1985兲.
5 S. Sugai, J. Phys. C 18, 799 共1985兲.
6
N. Hamaya, K. Sato, K. Usui-Watanabe, K. Fuchizaki, Y. Fujii,
and Y. Ohishi, Phys. Rev. Lett. 79, 4597 共1997兲.
7
S. Sharma and S. Sikka, Prog. Mater. Sci. 40, 1 共1996兲.
8 A. Pereira, C. A. Perottoni, and J. da Jornada, J. Raman Spectrosc. 34, 578 共2003兲.
9 O. Mishima, L. Calvert, and E. Whalley, Nature 共London兲 310,
393 共1984兲.
10 B. Riggleman and H. Drickamer, J. Chem. Phys. 38, 2721
共1963兲.
11
A. L. Chen, P. Y. Yu, and M. P. Pasternak, Phys. Rev. B 44, 2883
共1991兲.
12
D. Heinz and R. Jeanloz, J. Appl. Phys. 55, 885 共1984兲.
13 H. Mao, J. Xu, and P. Bell, J. Geophys. Res., 关Solid Earth Planets兴 91, 4673 共1986兲.
14 A. Ohmura, N. Hamaya, K. Sato, C. Ogawa, M. Isshiki, and Y.
1 R.
struction and calibration of the heating apparatus, and the
anonymous reviewers for helpful comments. We acknowledge the use of the facilities of beamline 12.2.2. at the Advanced Light Source, Lawrence Berkeley National Laboratory. The Advance Light Source is supported by the Director,
Office of Science, Office of Basic Energy Sciences, Materials Sciences Division, of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098 at Lawrence Berkeley National Laboratory.
Ohishi, J. Phys.: Condens. Matter 14, 10553 共2002兲.
Fuchizaki, S. Sugiyama, and Y. Fujii, J. Chem. Phys. 112,
10379 共2000兲.
16 R. Lynch and H. Drickamer, J. Chem. Phys. 45, 1020 共1966兲.
17
K. Catalli, S. Shim, and V. Prakapenka, Geophys. Res. Lett. 35,
L05312 共2008兲.
18 W. Caldwell, Ph.D. thesis, University of California–Berkeley,
2000.
19 K. J. Kingma, C. Meade, R. Hemley, H. Mao, and D. Veblen,
Science 259, 666 共1993兲.
20 K. Fuchizaki, S. Sugiyama, and Y. Fujii, J. Phys. Soc. Jpn. 70,
1321 共2001兲.
21 R. Hemley, A. Jephcoat, H. Mao, L. Ming, and M. Manghnani,
Nature 共London兲 334, 52 共1988兲.
22
K. Fuchizaki and K. Nagai, Solid State Commun. 132, 305
共2004兲.
23 K. Fuchizaki, Y. Fujii, A. Ohishi, N. Hamaya, Y. Katayama, and
T. Okada, J. Chem. Phys. 120, 11196 共2004兲.
24 M. Kruger and C. Meade, Phys. Rev. B 55, 1 共1997兲.
25 M. Kruger and R. Jeanloz, Science 249, 647 共1990兲.
26 J. S. Tse and D. D. Klug, Phys. Rev. Lett. 67, 3559 共1991兲.
15 K.
094101-7